The proposed project is a multi-method longitudinal field experiment to develop a comprehensive partner notification model targeting HIV-positive injecting drug users and their sexual and/or needle-sharing partners. The project evaluates the efficacy of combining community-based outreach and prevention activities with partner notification. The project will determine which of two models (index person only vs. outreach assisted) is more effective for motivating HIV positive injecting drug-users to disclose their sexual and needle partners and notify them of the need to be tested. Phase one of the project will explore the concerns and constraints that deter HIV positive injecting drug users from disclosing and notifying their needle and sex partners. Multiple investigative techniques will be used including focus groups, individual indepth interviews, and ethnographic field work to discern the barriers to notification. Because of the. extreme sensitivity associated with HIV partner notification, information from these sources will be used to design the final study protocols, notification and follow-up procedures, sample size, and human subjects concerns. This phase also will provide insight into the likely impact of notification on the psychological well-being, partner relationships, and drug network ties for both index persons and their partners. Phase two combines quantitative measures, HIV test-results, and ethnographic data to evaluate the efficacy of the index-person only vs. outreach-assisted models. Staff who are indigenous to the community will provide conventional AIDS outreach services that include AIDS prevention information and materials. 450 IDUs will be recruited for interview, pre- test counseling, and HIV testing. Following testing, subjects will be asked to return in two weeks for re-interview, the opportunity to learn their test results and post test counseling. HIV positive individuals will be defined as index subjects for the purposes of the study and randomly assigned to one of two notification model where: (1) the subject assumes sole responsibility for partner notification, or (2) the subject either selects sole responsibility or some combination of Outreach- assisted/personal notification. After four months, multiple contact methods will be used to assess index persons? compliance with their assigned notification model. Partners of index persons in the study will be offered pre-test-post counseling, HIV testing, and enrollment in the project as second generation subjects if they are HIV positive. Using the same procedures as with first generation index subjects, partners who test HIV-positive will be assigned to one of the two notification models being tested. Effects of testing and notification on the psychosocial well being, behavioral risk activities and interpersonal relationships among index persons and their partners also will be examined.